xlm-roberta-base-2-contract-sections-classification-v4-50

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2435
Accuracy Evaluate: 0.9653
Precision Evaluate: 0.9654
Recall Evaluate: 0.9677
F1 Evaluate: 0.9662
Accuracy Sklearn: 0.9653
Precision Sklearn: 0.9656
Recall Sklearn: 0.9653
F1 Sklearn: 0.9652
Acuracia Rotulo Objeto: 0.9855
Acuracia Rotulo Obrigacoes: 0.9646
Acuracia Rotulo Valor: 0.9398
Acuracia Rotulo Vigencia: 0.9816
Acuracia Rotulo Rescisao: 0.9640
Acuracia Rotulo Foro: 0.95
Acuracia Rotulo Reajuste: 0.9751
Acuracia Rotulo Fiscalizacao: 0.8927
Acuracia Rotulo Publicacao: 1.0
Acuracia Rotulo Pagamento: 0.9855
Acuracia Rotulo Casos Omissos: 0.9409
Acuracia Rotulo Sancoes: 1.0
Acuracia Rotulo Dotacao Orcamentaria: 1.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-06
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Evaluate	Precision Evaluate	Recall Evaluate	F1 Evaluate	Accuracy Sklearn	Precision Sklearn	Recall Sklearn	F1 Sklearn	Acuracia Rotulo Objeto	Acuracia Rotulo Obrigacoes	Acuracia Rotulo Valor	Acuracia Rotulo Vigencia	Acuracia Rotulo Rescisao	Acuracia Rotulo Foro	Acuracia Rotulo Reajuste	Acuracia Rotulo Fiscalizacao	Acuracia Rotulo Publicacao	Acuracia Rotulo Pagamento	Acuracia Rotulo Casos Omissos	Acuracia Rotulo Sancoes	Acuracia Rotulo Dotacao Orcamentaria
1.852	1.0	1000	1.5503	0.6855	0.7334	0.6913	0.6800	0.6855	0.7240	0.6855	0.6796	0.9587	0.6061	0.4842	0.6037	0.6925	0.9231	0.8221	0.2366	0.9015	0.8297	0.8325	0.7890	0.3077
0.875	2.0	2000	0.7078	0.88	0.8938	0.8952	0.8912	0.88	0.8879	0.88	0.8793	0.9628	0.7340	0.6991	0.9344	0.9224	0.9269	0.9466	0.8644	0.9655	0.8986	0.9064	0.8991	0.9780
0.4719	3.0	3000	0.3940	0.92	0.9267	0.9312	0.9277	0.92	0.9239	0.92	0.9202	0.9690	0.7997	0.8625	0.9685	0.9668	0.9385	0.9644	0.8738	0.9901	0.9239	0.9261	0.9450	0.9780
0.2712	4.0	4000	0.3128	0.9283	0.9349	0.9403	0.9364	0.9283	0.9319	0.9283	0.9285	0.9587	0.8047	0.9169	0.9764	0.9723	0.9385	0.9786	0.8738	1.0	0.9312	0.9310	0.9633	0.9780
0.159	5.0	5000	0.3166	0.9283	0.9368	0.9411	0.9374	0.9283	0.9330	0.9283	0.9285	0.9649	0.7896	0.9427	0.9764	0.9778	0.9385	0.9644	0.8770	1.0	0.9167	0.9360	0.9725	0.9780
0.1061	6.0	6000	0.2970	0.9307	0.9376	0.9444	0.9394	0.9307	0.9355	0.9307	0.9310	0.9649	0.7912	0.9284	0.9764	0.9778	0.9385	0.9858	0.8801	1.0	0.9384	0.9360	0.9817	0.9780
0.0936	7.0	7000	0.2725	0.9355	0.9433	0.9479	0.9445	0.9355	0.9381	0.9355	0.9353	0.9731	0.8098	0.9398	0.9764	0.9751	0.9385	0.9893	0.8801	1.0	0.9312	0.9409	0.9908	0.9780
0.0781	8.0	8000	0.2621	0.9383	0.9432	0.9514	0.9459	0.9383	0.9422	0.9383	0.9385	0.9731	0.8114	0.9284	0.9816	0.9695	0.9385	0.9858	0.8864	1.0	0.9746	0.9409	1.0	0.9780
0.0664	9.0	9000	0.2128	0.9553	0.9563	0.9621	0.9589	0.9553	0.9555	0.9553	0.9552	0.9711	0.9057	0.9312	0.9816	0.9640	0.9962	0.9893	0.8833	1.0	0.9746	0.9409	0.9908	0.9780
0.0616	10.0	10000	0.2505	0.948	0.9493	0.9590	0.9530	0.948	0.9507	0.948	0.9483	0.9752	0.8485	0.9513	0.9843	0.9695	0.95	0.9858	0.8864	1.0	0.9746	0.9409	1.0	1.0
0.0687	11.0	11000	0.1945	0.9663	0.9667	0.9681	0.9669	0.9663	0.9668	0.9663	0.9662	0.9793	0.9630	0.9456	0.9843	0.9751	0.9846	0.9964	0.8833	1.0	0.9638	0.9409	0.9908	0.9780
0.0532	12.0	12000	0.1884	0.965	0.9650	0.9696	0.9670	0.965	0.9652	0.965	0.9649	0.9876	0.9343	0.9513	0.9843	0.9751	0.9846	0.9893	0.8833	1.0	0.9746	0.9409	1.0	1.0
0.0658	13.0	13000	0.1943	0.963	0.9672	0.9667	0.9667	0.963	0.9634	0.963	0.9629	0.9876	0.9394	0.9542	0.9843	0.9668	0.9846	0.9893	0.8833	1.0	0.9457	0.9409	0.9908	1.0
0.0415	14.0	14000	0.1836	0.9698	0.9702	0.9721	0.9707	0.9698	0.9701	0.9698	0.9696	0.9876	0.9663	0.9513	0.9843	0.9751	0.9962	0.9822	0.8833	1.0	0.9746	0.9360	1.0	1.0
0.0414	15.0	15000	0.2002	0.9683	0.9679	0.9711	0.9689	0.9683	0.9688	0.9683	0.9681	0.9855	0.9697	0.9226	0.9843	0.9612	1.0	0.9964	0.8801	1.0	0.9891	0.9409	1.0	0.9945
0.0329	16.0	16000	0.1970	0.967	0.9683	0.9702	0.9687	0.967	0.9676	0.967	0.9669	0.9876	0.9680	0.9398	0.9816	0.9363	0.9962	0.9964	0.8801	1.0	0.9855	0.9409	1.0	1.0
0.0382	17.0	17000	0.1987	0.9695	0.9704	0.9718	0.9706	0.9695	0.9699	0.9695	0.9693	0.9855	0.9731	0.9370	0.9843	0.9695	0.9962	0.9929	0.8801	1.0	0.9746	0.9409	1.0	1.0
0.036	18.0	18000	0.1994	0.9673	0.9688	0.9704	0.9692	0.9673	0.9676	0.9673	0.9671	0.9917	0.9596	0.9312	0.9816	0.9695	0.9962	0.9751	0.8833	1.0	0.9855	0.9409	1.0	1.0
0.0399	19.0	19000	0.2332	0.9565	0.9636	0.9586	0.9602	0.9565	0.9589	0.9565	0.9567	0.9917	0.9495	0.9427	0.9816	0.9557	0.9962	0.8505	0.8864	1.0	0.9855	0.9409	0.9817	1.0
0.0307	20.0	20000	0.2063	0.9663	0.9710	0.9686	0.9694	0.9663	0.9669	0.9663	0.9662	0.9876	0.9663	0.9427	0.9816	0.9723	1.0	0.9751	0.8833	1.0	0.9420	0.9409	1.0	1.0
0.0256	21.0	21000	0.2059	0.9667	0.9712	0.9696	0.9700	0.9667	0.9675	0.9667	0.9667	0.9897	0.9596	0.9456	0.9816	0.9584	0.9962	0.9680	0.8833	1.0	0.9855	0.9458	0.9908	1.0
0.0231	22.0	22000	0.1974	0.9692	0.9707	0.9721	0.9709	0.9692	0.9698	0.9692	0.9691	0.9876	0.9663	0.9513	0.9843	0.9612	0.9962	0.9786	0.8801	1.0	0.9855	0.9458	1.0	1.0
0.0244	23.0	23000	0.2056	0.9677	0.9691	0.9707	0.9695	0.9677	0.9681	0.9677	0.9676	0.9855	0.9663	0.9284	0.9816	0.9751	0.9962	0.9786	0.8801	1.0	0.9819	0.9458	1.0	1.0
0.0296	24.0	24000	0.2062	0.9698	0.9702	0.9728	0.9712	0.9698	0.9701	0.9698	0.9697	0.9876	0.9596	0.9484	0.9816	0.9695	0.9962	0.9715	0.9085	1.0	0.9819	0.9409	1.0	1.0
0.0301	25.0	25000	0.2194	0.9673	0.9687	0.9703	0.9691	0.9673	0.9678	0.9673	0.9672	0.9835	0.9646	0.9456	0.9843	0.9584	0.9962	0.9715	0.8833	1.0	0.9855	0.9409	1.0	1.0
0.0189	26.0	26000	0.2081	0.9688	0.9687	0.9717	0.9697	0.9688	0.9693	0.9688	0.9687	0.9855	0.9630	0.9542	0.9816	0.9640	1.0	0.9786	0.8896	1.0	0.9746	0.9409	1.0	1.0
0.0186	27.0	27000	0.2231	0.962	0.9679	0.9641	0.9652	0.962	0.9638	0.962	0.9620	0.9917	0.9545	0.9570	0.9816	0.9861	0.9962	0.8577	0.8896	1.0	0.9783	0.9409	1.0	1.0
0.0201	28.0	28000	0.2071	0.9695	0.9725	0.9718	0.9719	0.9695	0.9699	0.9695	0.9694	0.9855	0.9646	0.9456	0.9843	0.9723	0.9962	0.9751	0.8927	1.0	0.9855	0.9409	0.9908	1.0
0.0202	29.0	29000	0.2120	0.9688	0.9700	0.9708	0.9701	0.9688	0.9690	0.9688	0.9687	0.9855	0.9646	0.9456	0.9843	0.9778	0.9846	0.9680	0.8927	1.0	0.9855	0.9409	0.9908	1.0
0.0201	30.0	30000	0.2111	0.9702	0.9734	0.9722	0.9725	0.9702	0.9706	0.9702	0.9701	0.9917	0.9630	0.9427	0.9843	0.9889	0.9962	0.9715	0.8833	1.0	0.9855	0.9409	0.9908	1.0
0.0181	31.0	31000	0.2263	0.9683	0.9685	0.9713	0.9695	0.9683	0.9688	0.9683	0.9682	0.9835	0.9646	0.9484	0.9843	0.9584	0.9962	0.9786	0.8864	1.0	0.9855	0.9409	1.0	1.0
0.0145	32.0	32000	0.2255	0.9645	0.9655	0.9658	0.9653	0.9645	0.9651	0.9645	0.9645	0.9814	0.9646	0.9542	0.9843	0.9668	0.9385	0.9715	0.8864	1.0	0.9855	0.9409	0.9817	1.0
0.0192	33.0	33000	0.2265	0.966	0.9675	0.9687	0.9676	0.966	0.9666	0.966	0.9659	0.9835	0.9663	0.9398	0.9790	0.9668	0.9962	0.9680	0.8801	1.0	0.9819	0.9310	1.0	1.0
0.0121	34.0	34000	0.2339	0.9685	0.9686	0.9714	0.9697	0.9685	0.9691	0.9685	0.9685	0.9835	0.9646	0.9456	0.9816	0.9584	0.9962	0.9644	0.9117	1.0	0.9819	0.9409	1.0	1.0
0.0179	35.0	35000	0.2327	0.9683	0.9685	0.9707	0.9693	0.9683	0.9687	0.9683	0.9682	0.9835	0.9646	0.9513	0.9816	0.9640	0.9846	0.9680	0.9054	1.0	0.9855	0.9310	1.0	1.0
0.0084	36.0	36000	0.2162	0.9688	0.9720	0.9705	0.9709	0.9688	0.9693	0.9688	0.9687	0.9897	0.9579	0.9570	0.9843	0.9778	0.9962	0.9786	0.8959	1.0	0.9565	0.9409	0.9817	1.0
0.0115	37.0	37000	0.2436	0.9645	0.9649	0.9668	0.9654	0.9645	0.9648	0.9645	0.9644	0.9835	0.9663	0.9370	0.9790	0.9695	0.9577	0.9751	0.8833	1.0	0.9855	0.9310	1.0	1.0
0.0146	38.0	38000	0.2464	0.9647	0.9684	0.9672	0.9673	0.9647	0.9656	0.9647	0.9647	0.9855	0.9646	0.9456	0.9816	0.9363	0.9962	0.9751	0.8801	1.0	0.9855	0.9409	0.9817	1.0
0.0112	39.0	39000	0.2432	0.9635	0.9637	0.9659	0.9644	0.9635	0.9641	0.9635	0.9635	0.9835	0.9646	0.9456	0.9816	0.9557	0.9423	0.9680	0.8927	1.0	0.9819	0.9409	1.0	1.0
0.011	40.0	40000	0.2498	0.9627	0.9627	0.9652	0.9636	0.9627	0.9634	0.9627	0.9627	0.9835	0.9630	0.9484	0.9816	0.9557	0.9385	0.9573	0.8927	1.0	0.9855	0.9409	1.0	1.0
0.0124	41.0	41000	0.2484	0.964	0.9640	0.9667	0.9649	0.964	0.9644	0.964	0.9639	0.9835	0.9646	0.9398	0.9790	0.9584	0.95	0.9751	0.8896	1.0	0.9855	0.9409	1.0	1.0
0.011	42.0	42000	0.2475	0.9645	0.9669	0.9666	0.9664	0.9645	0.9651	0.9645	0.9645	0.9855	0.9646	0.9484	0.9843	0.9474	0.9577	0.9680	0.8927	1.0	0.9855	0.9409	0.9908	1.0
0.0107	43.0	43000	0.2413	0.9645	0.9658	0.9664	0.9658	0.9645	0.9649	0.9645	0.9644	0.9855	0.9630	0.9456	0.9843	0.9612	0.9385	0.9751	0.8927	1.0	0.9855	0.9409	0.9908	1.0
0.0078	44.0	44000	0.2472	0.9637	0.9652	0.9664	0.9655	0.9637	0.9642	0.9637	0.9637	0.9855	0.9646	0.9398	0.9790	0.9529	0.95	0.9751	0.8896	1.0	0.9855	0.9409	1.0	1.0
0.0134	45.0	45000	0.2475	0.9635	0.9669	0.9654	0.9658	0.9635	0.9642	0.9635	0.9635	0.9855	0.9646	0.9456	0.9816	0.9529	0.9385	0.9715	0.8927	1.0	0.9855	0.9409	0.9908	1.0
0.0092	46.0	46000	0.2444	0.9643	0.9655	0.9666	0.9657	0.9643	0.9647	0.9643	0.9642	0.9855	0.9646	0.9398	0.9790	0.9640	0.9385	0.9751	0.8927	1.0	0.9855	0.9409	1.0	1.0
0.0083	47.0	47000	0.2452	0.9647	0.9660	0.9670	0.9661	0.9647	0.9652	0.9647	0.9647	0.9855	0.9646	0.9427	0.9843	0.9612	0.9385	0.9751	0.8927	1.0	0.9855	0.9409	1.0	1.0
0.008	48.0	48000	0.2429	0.9653	0.9650	0.9677	0.9660	0.9653	0.9656	0.9653	0.9652	0.9855	0.9646	0.9398	0.9816	0.9640	0.95	0.9751	0.8927	1.0	0.9855	0.9409	1.0	1.0
0.0078	49.0	49000	0.2429	0.9653	0.9652	0.9677	0.9661	0.9653	0.9655	0.9653	0.9652	0.9855	0.9646	0.9398	0.9816	0.9640	0.95	0.9751	0.8927	1.0	0.9855	0.9409	1.0	1.0
0.0086	50.0	50000	0.2435	0.9653	0.9654	0.9677	0.9662	0.9653	0.9656	0.9653	0.9652	0.9855	0.9646	0.9398	0.9816	0.9640	0.95	0.9751	0.8927	1.0	0.9855	0.9409	1.0	1.0

Framework versions

Transformers 4.48.3
Pytorch 2.6.0+cu124
Datasets 3.3.0
Tokenizers 0.21.0

marcelovidigal/xlm-roberta-base-2-contract-sections-classification-v4-50

xlm-roberta-base-2-contract-sections-classification-v4-50

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种